Designs, builds, and owns production RAG systems, including ingestion, chunking, embeddings, hybrid retrieval, reranking, evaluation, secure multi-tenant access, and LLM serving. Optimizes open-source models with Ollama and llama.cpp across cloud, on-premises, and air-gapped environments. Establishes architecture standards, reviews designs, mentors AI developers, evaluates emerging RAG techniques, and supports Azure/AWS infrastructure, Kubernetes deployments, CI/CD, observability, and Kafka-based ingestion pipelines.

Build and operate production LLM systems using Ollama and llama.cpp, develop RAG pipelines and MCP integrations, and create Python backend services for AI features. Deploy and maintain workloads across Azure and AWS using Docker and CI/CD. Automate operations, process structured data, support Kafka-based architectures, troubleshoot across the stack, and progressively develop expertise in model quantization, GPU tuning, and secure offline deployments.